Differences

This shows you the differences between two versions of the page.

--- development:simd [2025/10/30] – created hayati
+++ development:simd [Unknown date] (current) – external edit (Unknown date) 127.0.0.1
@@ Line 1: / Line 1: @@
+~~NOTRANS~~
+{{indexmenu_n>400}}
+====== SIMD ======
+Modern processors have capabilities for SIMD (single instruction multiple data) like SSE or AVX instruction sets on x64 or NEON on ARM architecture. Now, in 2022, C/C++ compilers still have problems at automatic vectorization to generate these vector instructions. Assembler or compiler intrinsic functions can be used to enforce the usage ..
+===== CPU architectures/infos =====
+  * https://en.wikichip.org/wiki/x86/extensions
+  * https://en.wikichip.org/wiki/arm/versions
+  * https://en.wikichip.org/wiki/arm_holdings#Microarchitectures
+  * https://en.wikipedia.org/wiki/Raspberry_Pi#Specifications
+    * https://en.wikipedia.org/wiki/Comparison_of_ARMv8-A_processors
+    * does contain the L1/L2/L3 cache sizes
+    * Raspberry Pi 4 has Broadcom Chip BCM2711
+===== Libraries =====
+  * https://www.intel.com/content/www/us/en/docs/intrinsics-guide/index.html
+  * https://github.com/VcDevel/Vc
+  * https://github.com/hayguen/MIPP
+    * it's origin https://github.com/aff3ct/MIPP
+  * SSE/AVX to Neon
+    * https://github.com/DLTcollab/sse2neon
+    * https://github.com/kunpengcompute/AvxToNeon
+  * Neon for x86/SSE
+    * https://github.com/intel/ARM_NEON_2_x86_SSE
+  * SIMD Everywhere
+    * https://simd-everywhere.github.io/blog/
+    * https://github.com/simd-everywhere/simde
+    * https://github.com/jpcima/simde